Computer Science 252 Problem Solving with Java The College of Saint Rose Spring Topic Notes: Collections

Transcription

1 Computer Science 252 Problem Solving with Java The College of Saint Rose Spring 2014 Topic Notes: Collections Our next major topic involves additional common mechanisms for naming collections of items. Motivation for Collections Sometimes we have a lot of very similar data, and we would like to do similar things to each datum. For example, suppose we wanted to extend our Drag2Shirts example to have 4 shirts instead of just 2. See Example: Drag2Shirts We could go through the program and everywhere we see redshirt and blueshirt, add 2 more variables and 2 more segments of code to deal with the new 2 shirts. See Example: Drag4Shirts That was not terribly painful, but a bit tedious and error prone. Now, what if we wanted to create 10, 20, or 100 shirts to be dragged around the canvas. We d want a better way to name the shirts as a group. We could create a recursive data structure to hold our TShirt objects like we did in some recent examples. This is a useful approach in many cases, but we will now consider some other very common ways to manage collections of objects. Java and other programming languages provide a number of mechanisms to help here. We will consider two in Java. First, we will look at a Java class called the ArrayList, and later a lowerlevel construct common to most modern programming language called arrays. Each allows us to use one name for an entire collection of objects. The Java ArrayList Class Those of you who will go on to take data structures will learn about a variety of ways that collections of data can be stored that vary in complexity, flexibility, and efficiency. We will consider just one of those structures here: the ArrayList. ArrayList is a class that implements an abstract data type provided by the standard Java utility library. Let s see how to use them through an example: we will replace the 4 names of TShirt objects in the Drag4Shirts example with a single ArrayList that holds all 4. See Example: Drag4ShirtsArrayList

2 This program has the same functionality, but the 4 variables for the TShirts has been replaced by a single collection, an ArrayList of TShirt objects. We consider each change that was made to the program to see the basic usage of an ArrayList. First, we need to add an import statement to the top of our program. import java.util.arraylist; This allows us to use the class name ArrayList in the rest of the file and Java will know we mean to use the one in the java.util package. Next, we declare an instance variable for our ArrayList: private ArrayList<TShirt> shirts; This looks a little different than any variable declaration we have seen before. Since an ArrayList can be used to hold objects of any type, we need to tell Java what type of objects will be stored in this particular ArrayList. In this case, it s TShirts. So we place that type inside the < and >. This is called a type parameter. Like most Java classes, we need to construct an instance of the class in order use it. This is done in the first statement of the begin method: shirts = new ArrayList<TShirt>(); This is much like other constructions we have seen, but we again need to include the type parameter so Java will give us an ArrayList that is set up to hold a collection of TShirt objects. The TShirt instances are then created, and we need to insert each into the ArrayList. This is done with the add method: shirts.add(shirt); This will take the TShirt named shirt and add it to the first available slot in the ArrayList named shirts. Note that in this case, we are not requesting any specific location within the ArrayList for the shirt. We will later see that we can be more specific here. Note also that we as users of the ArrayList do not know (though when you take data structures, you ll have a pretty good idea) of what s going on inside the ArrayList to add the shirt. We just know that it knows how to do it. When we re done with begin, the ArrayList contains references to 4 TShirt objects. 2

3 In the onmousepress and onmouseexit methods, we need to access the TShirt objects within the ArrayList. We do this with the get method: TShirt shirt = shirts.get(shirtnum); Here, shirtnum is a loop index variable that will range from 0 to one less than the number of items in the ArrayList. We know in this case that there are 4 items, but we can get that information from the ArrayList itself using the size method, as done in the for loops: for (int shirtnum = 0; shirtnum < shirts.size(); shirtnum++) What we see here is that the ArrayList has assigned a number, often called an index, to each TShirt we added to the ArrayList, and we can pass that number to the get method to get back a specific TShirt from the ArrayList. It turns out that the first item we add is given index 0, the next is given index 1, and so on. If we later wanted to get at the first one, we could say: shirts.get(0); but in many cases (like this one), we will access the items within a collection inside a loop, passing in a loop index to the get method. This is our first example of a search operation on a collection we are looking through each object in the collection to find one that contains the Location. More precisely, this is a linear search and we will say more about this later. One of the great things about using a construct like an ArrayList is that we can extend our programs to keep track of a much larger number of objects. If we want to have 10 TShirts on the canvas, we would definitely want to use a collection like an ArrayList to keep track of them. See Example: Drag10Shirts Here, we also place the creation of the TShirts into a loop, but just line them up in a row for simplicity. If we wanted them to be organized into rows or to use a fixed set of colors, we would need to use a more complicated loop in the begin method. (And we will do just that later.) If we wanted to create 20 or 50 or 100 Tshirts, we could do so by changing the loop in the begin method and the remainder of the code does not need to change. ArrayLists in Custom Objects One of the challenges we have seen with constructing custom objects with any level of complexity is that we need to have names for all of the graphical objects we construct. When the object includes large numbers of items, ideally created within a loop, an ArrayList will come in handy to help keep track of them. 3

4 First, we look at a program that doesn t use ArrayLists: See Example: DrawRoads This program draws little segments of roads when we click the mouse. Nothing is new here we could have written this a while ago. But now suppose we want to be able to drag one of these around. We need to have names for all of the components of the road segment so we can do things like move it and check for containment of a point. See Example: DragRoads The enhancements to the WindowController class are all very familiar. It s in the RoadSegment class that we make use of an ArrayList to hold the center stripes of our road segment. Notice the same steps: declare a variable with an ArrayList type that can hold objects of the appropriate type, construct it with new, then add entries with the appropriate types of objects. In the constructor, we do the construction of the ArrayList, then create the actual stripes. In the move method, we loop through the stripes, moving each one. This is nice, but perhaps we want to combine this functionality with that of the program where we could drag around any of 10 shirts. Let s use an ArrayList to keep track of all of the road segments we ve created, so we can drag any segment, not just the most recently drawn one. See Example: DragAllRoads Here, in addition to having an ArrayList to keep track of the components of one of the road segments, we keep an ArrayList of RoadSegment objects in the WindowController class. Removing from an ArrayList We can augment the last example to remove each road segment from the canvas and from the ArrayList. A road segment will be removed if it is being dragged when the mouse leaves the window. See Example: DragAllRoadsRemove The new functionality is in the onmouseexit method of the DragAllRoadsRemove class. If the dragging flag is true when the mouse leaves the window, the currently-dragged segment selectedsegment) should be removed. We first remove it from the canvas, then remove it from the ArrayList. We also set dragging back to false, since the object we were just dragging no longer exists. First, we will look at the removal from the list, which is done with the ArrayList s remove method. We pass as a parameter the element we want to remove, and if it is an element of the list, it is removed. It is important to note that when we remove an element from an ArrayList 4

5 with remove, any subsequent entries will be moved up. That is, if a list contains 5 elements (in positions numbered 0 through 4) and we remove the element at position 2, the ArrayList implementation of remove will shift the element that was in position 3 into position 2, and the one that was in position 4 into position 3. This means we can still use our for loop over the numbers from 0 to size()-1 to visit all of our entries. In other words, remove does not leave a hole at the index from which the element was removed. The new removefromcanvas method is mostly like the ones we have seen in previous examples: to remove the custom object, we remove each of its components. The difference here is that we need to loop through the ArrayList, get and then remove each element. We also should remove the individual FilledRects from the ArrayList, which we do all at once with the clear method. We can also remove elements from an ArrayList by index rather than value. We will see examples of this soon. Other ArrayList methods The examples above demonstrated just a few of the capabilities of the ArrayList class: construction, add, get, size, remove, and clear. The full documentation for the ArrayList can be found at javase/6/docs/api/java/util/arraylist.html Here are a couple of additional methods, some of which will come up in later examples. contains determine if a given object is in the list indexof search for first occurrence of a given object in the list and return its index set replace the contents at an index with a new element A few more examples to bring some of this together: See Example: MovingFlags See Example: PongBricks ArrayLists of Primitive Types Java places a significant restriction on the use of primitive types as the type parameters for generic data structures such as the ArrayList. The following would not be valid Java: ArrayList<int> a = new ArrayList<int>(); The type in the <> must be an object type. Fortunately, Java provides object types that correspond to each primitive type. An Integer object is able to store a single int value, a Double value is able to store a single double value, etc. So the declaration and construction above would need to be: 5

6 ArrayList<Integer> a = new ArrayList<Integer>(); In older versions of Java, programmers would need to be careful to convert back and forth between values of the primitive types and their object encapsulators. To construct an Integer from an int i, one would need to do so explicitly: a.add(new Integer(i)); And to retrieve the int value from an Integer, one would also do so explicitly: a.get(pos).intvalue(); However, recent versions of Java automatically convert between the primitive types and their object encapsulating classes. This is called autoboxing when converting from primitive to boxed encapsulating classes, and autounboxing when going back the other way. However, the effective programmer should always keep in mind that these conversions are occurring, as there is a computational cost to each. Another Example Suppose we have an ArrayList of Integer values, and someone (by a mechanism which is not our concern) has asked us to write a method that will find the largest value in the ArrayList. The following method will achieve this (we assume at least one element in the ArrayList): private static int findmax(arraylist<integer> a) { int max = a.get(0); for (int i=1; i<a.size(); i++) { int val = a.get(i); if (val > max) max = val; return max; The for-each Loop We have seen that a common task with a collection such as an ArrayList is to iterate over its contents. That is, visit every element in the list exactly once to do something to it. It is often the case (and was in many of the examples here) that the specific index of an entry in an ArrayList is not important as we are iterating over its contents. For example, consider the See Example: DragAllRoadsRemove example. It doesn t actually matter the order in which 6

7 we process the items in the ArrayLists in the move and removefromcanvas methods of RoadSegment or in the onmousepress method of DragAllRoadsRemove. In these cases, the counting for loops can be replaced with a related Java construct often called the for-each loop. If we have an ArrayList of objects of some type T and we wish to loop over all entries in the loop, we can replace a counting loop: ArrayList<T> a = new ArrayList<T>();... for (int i = 0; i < a.size(); i++) { T item = a.get(i); // do something with item with a for-each loop: ArrayList<T> a = new ArrayList<T>();... for (T item : a) { // do something with item This construct will loop enough times so that the variable item will be assigned to each entry in a exactly once through the body of the loop. See Example: DragAllRoadsForeach The for-each construct is not always appropriate, however. For example, in the findmax method above, it is more convenient to be able to get the item at position 0 as the initial max and then loop over the entries from positions 1 and up to check for larger values. As you learn more Java, you will see a number of other data structures that can be used with the for-each loop construct. Java Arrays The ArrayList is a Java class, provided as a standard utility with every Java environment. But it is built on top of a more fundamental programming language construct called an array. 7

8 In mathematics, we can refer to large groups of numbers (for example) by attaching subscripts to names. We can talk about numbers n 1, n 2,... An array lets us do the same thing with computer languages. Suppose we wish to have a group of elements all of which have type ThingAMaJig and we wish to call the group things. Then we write the declaration of things as ThingAMaJig[] things; The only difference between this and the declaration of a single item of type ThingAMaJig is the occurrence of [ ] after the type. Like all other objects, a group of elements needs to be created: things = new ThingAMaJig[25]; Again, notice the square brackets. The number in parentheses (25) indicates the number of slots to create, each of which can hold one of the elements. We can now refer to individual elements using subscripts. However, in programming languages we cannot easily set the subscripts in a smaller font placed slightly lower than regular type. As a result we use the ubiquitous [ ] to indicate a subscript. If, as above, we define things to have 25 elements, they may be referred to as: things[0], things[1],..., things[24] We start numbering the subscripts at 0, and hence the last subscript is one smaller than the total number of elements. Thus in the example above the subscripts go from 0 to 24. One warning: When we initialize an array as above, we only create slots for all of the elements, we do not necessarily fill the slots with elements. Actually, the default values of the elements of the array are the same as for instance variables of the same type. If ThingAMaJig is an object type, then the initial values of all elements is null, while if it is int, then the initial values will all be 0. Thus you will want to be careful to put the appropriate values in the array before using them (especially before sending message to them! that s a NullPointerException waiting to happen). In many ways, and array works like an ArrayList, but we will see several differences. Armed with this new construct, let s revisit the shirt dragging program to use arrays. See Example: Drag10ShirtsArray In this code, we we have a single array named shirts. This array is declared as an instance variable, constructed at the start of the begin method, and given values (references to actual TShirts) just after. Then in the onmousepress method, we loop through all of the array entries (as we did previously with an ArrayList) to determine which, if any, has been pressed. Finally, in onmouseexit, we tell all of the shirts to move back to their starting positions. 8

9 Let s see how this differs from the ArrayList version. Our instance variable declaration looks a bit different. When we construct the array in the begin method, we need to tell it how many elements the array will hold (in this case, 10). With the ArrayList, we construct a list and we can add as many things to it as we want. The array can only ever hold the number of elements we provided when we constructed it. When we add items to the array, we need to specify the index explicitly. There is no way to say just add it to the end the way we do with ArrayLists. When we access array elements, we use the bracket notation in much the same way we use the get method of the ArrayList. In this example, we have used an array to keep track of a collection of objects on the canvas. We can also use an array to keep track of the components of a custom object as we did with ArrayLists. An enhancement to this example that shows some of the benefits of arrays, we draw the t-shirts in two rows and use a fixed array of colors for the shirts instead of random colors. See Example: Drag10ShirtsNicer A few things to notice here: We have an array of Colors initialized to 10 pre-defined color names that we ll use for our 10 t-shirts. The construction of the t-shirts takes place in a nested loop to make it easier to organize them into 2 rows of 5 shirts each. Our next enhancement to this example is to draw and drag around 20 shirts, now in 4 rows of 5. See Example: Drag20Shirts Most of the program works correctly just by changing the value of the constant NUM ROWS (hooray for constants!). But...the array of colors is not large enough. We account for this by reusing the colors once we ve run out. This is accomplished with some modulo arithmetic: shirts[shirtnum].setcolor(shirtcolors[shirtnum % shirtcolors.length]); Another Example See Example: DragStudentsS14 9

10 What you ve been waiting for: being the stars of a program. This is another drag objects around example, but this time the objects being dragged are your pictures. In this example, we place the objects randomly on the canvas, but take some care to make sure they do not overlap at all. Notice the helper method overlapsany that helps ensure this. Any image being dragged is also made larger while it s being dragged. Other than that, it s similar to dragging 10 shirts. Arrays of Non-graphical Types There is no reason to limit our usage of arrays to graphical object types. The following is an example of a Java application (rather than an Applet it starts with a main method instead of begin and has no graphics canvas) that uses arrays of String, double, and int. See Example: GradeRangeCounter There are a few items here we haven t used much this semester (the Scanner) but which you have seen before. There are also examples of arrays declared and initialized as final, and an example of an array of int allocated with new. Inserting and Removing with Arrays We have already seen that there is quite a bit to keep track of when using arrays, especially when objects are being added. We need to manage both the size of the array and the number of items it contains. If it fills, we either need to make sure we do not attempt to add another element, or reconstruct the array with a larger size. As a wrapup of our initial discussion of arrays, let s consider two more situations and how we need to deal with them: adding a new item in the middle of an array, and removing an item from the end. For these examples, we will not use graphical objects, just numbers. Arrays can store numbers just as well as they can store references to objects. Suppose we have an array of int large enough to hold 20 numbers. The array would be declared as an instance variable: private int[] a; along with another instance variable indicating the number of ints currently stored in a: private int count; and constructed and initialized: 10

11 a = new int[20]; count = 0; At some point in the program, count contains 10, meaning that elements 0 through 9 of a contain meaningful values. Now, suppose we want to add a new item to the array. So far, we have done something like this: a[count] = 17; count++; This will put a 17 into element 10, and increment the count to 11. But suppose that instead, we want to put the 17 into element 5, and without overwriting any of the data currently in the array. Perhaps the array is maintaining the numbers in order from smallest to largest. In this case, we d first need to move up all of the elements in positions 5 through 9 to instead be in positions 6 through 10, add the 17 to position 5, and then increment count. If the variable insertat contains the position at which we wish to add a new value, and that new value is in the variable val: for (int i=count; i>insertat; i--) { a[i] = a[i-1] a[insertat] = val; count++; Now, suppose we would like to remove a value in the middle. Instead of moving up values to make space, we need to move down the values to fill in the hole that would be left by removing the value. If the variable removeat contains the index of the value to be removed: for (int i=removeat+1; i<count; i++) { a[i-1] = a[i]; count--; The loop is only necessary if we wish to maintain relative order among the remaining items in the array. If that is not important (as is often the case with our graphical objects), we might simply write: a[removeat] = a[count-1]; count--; 11

12 In circumstances where we are likely to insert or remove into the middle of an array during its lifetime, it usually makes sense to take advantage of the higher-level functionality of the ArrayList. Array and ArrayList Summary The following list summarizes the key differences and similarities between arrays and ArrayLists. Declaration To declare an array of elements of some type T: T[] ar; where T can be any type, including primitive types or Object types. And to declare an ArrayList that can hold items of type T: ArrayList<T> al; where T must be an object type. If we want to store a primitive type, we must use Java s corresponding object wrappers (e.g., Integer when we want to store items of type int). Construction To construct (allocate space for) our array of n elements of type T: ar = new T[n]; Once constructed, the array will always have space for n elements of type T if we want a larger or smaller array, we would have to construct a new one. The array constructed will have the default value for the datatype stored in each entry. For object types, all entries begin as null. For primitive number types, they begin as 0. For boolean arrays, they begin as false. To construct an ArrayList: al = new ArrayList<T>(); This ArrayList initially does not contain any values. Its size will be determined by the number of elements we add to it. Adding an Element To add an element to an array, we have to specify the position at which we wish to add the new element: ar[i] = t; 12

13 This will place the item t at position i into our array. i must be in the range 0 to n-1 if we constructed our array to have n entries. If there was already some data stored in position i, it will be overwritten with t. If we want to add the item to the end of the array, that is, the first unoccupied slot in the array, we will need an additional variable to keep track of the number of currently-occupied slots. If this is called asize, and we have been careful to make sure the asize elements in the array occupy slots 0 through asize-1, we can add the element with: ar[asize] = t; asize++; With an ArrayList, the add method takes care of this: al.add(t); Retrieving an Element To get an item from an array, we use the same notation. To put the value from position i in the array into some variable t: t = ar[i]; Whereas with the ArrayList, we need to call a method: t = al.get(i); Visiting All Elements To loop over all elements in the array: for (int i=0; i<asize; i++) { t = ar[i]; // do something with t and an ArrayList; for (int i=0; i<al.size(); i++) { t = al.get(i); // do something with t In both cases, we can also use the for-each loop. 13

14 Two-Dimensional Arrays We can create arrays to hold objects of any type, either basic data types like int and double, or instances of objects such as Image and FilledOval or TShirt. Nothing stops us from defining arrays of arrays. To declare an array, each of whose elements is an array of int: int[][] twodarray; While it is normally written without parentheses, we can think of the above declaration as defining twodarray as having type (int []) []. Thus each element of twodarray is an array of ints. Despite the fact that Java will treat this as an array of arrays, we usually think about this as a twodimensional array, with the elements arranged in a two-dimensional table so that twodarray[i][j] can be seen as the element in theith row andjth column. For example here is the layout for a twodimensional array a with 6 rows (numbered 0 to 5) and 4 columns: a[0][0] a[0][1] a[0][2] a[0][3] 1 a[1][0] a[1][1] a[1][2] a[1][3] 2 a[2][0] a[2][1] a[2][2] a[2][3] 3 a[3][0] a[3][1] a[3][2] a[3][3] 4 a[4][0] a[4][1] a[4][2] a[4][3] 5 a[5][0] a[5][1] a[5][2] a[5][3] Viewed in this way, our two-dimensional array is a grid, much like a map or a spreadsheet. This is a natural way to store things like tables of data or matrices. We access elements of two-dimensional arrays in a manner similar to that used for one dimensional arrays, except that we must provide both the row and column to access an element, giving the row number first. We create a two-dimensional array by providing the number of rows and columns. Thus we can create the two-dimensional array above by writing: int[][] a = new int[6][4]; (Though as good programmers, you would define constants for the number of rows and the number of columns.) A nested for loop is the most common way to access or update the elements of a two-dimensional array. One loop walks through the rows and the other walks through the columns. For example, if we wanted to assign a unique number to each cell of our two-dimensional array, we could do the following: 14

15 for (int row = 0; row < 6; row++) { for (int col = 0; col < 4; col++) { a[row][col] = 4*row + col + 1; This assigns the numbers 1 through 24 to the elements of array a. The array is filled by assigning values to the elements in the first row, then the second row, etc. and results in: And if we wanted to print the above, we can write a loop: for (int row = 0; row < 6; row++) { for (int col = 0; col < 4; col++) { System.out.print(a[row][col] + " "); System.out.println(); You could modify the above to be slightly more interesting by computing a multiplication table. We could just as well process all the elements of column 0 first, then all of column 1, etc., by swapping the order of our loops: for (int col = 0; col < 4; col++) for (int row = 0; row < 6; row++)... For the most part, it doesn t matter which order you choose, though for large arrays it is generally a good idea to traverse the array in the same order that your programming language will store the values in memory. For Java (and C, C++), the data is stored by rows, known as row major order. However, an two-dimensional array in FORTRAN is stored in column major order. You will almost certainly see this again if you go on and take courses like Computer Organization or Operating Systems. Let s look at an example that makes use of a small (3 3) two-dimensional array: a tic-tac-toe game. 15

16 See Example: TicTacToe Most of the program is pretty straightforward, so we ll focus on the use of the two-dimensional arrays and discuss some of the private helper methods that make the code simpler, especially when checking for winning boards. The array that represents the board is a two-dimensional array of int called marks. Each entry will contain one of three values that will indicate if the cell is empty (0), contains an X (1) or contains an O (2). Named constants make the code involving these numbers easier to understand. Several private helper methods are provided to draw the board and the X s and O s. Look a bit at the onmousepress method, where the game is played. When the mouse is pressed and it is determined which (if any) of the 9 spaces contain the press point, we check to see if the space is occupied. If not, the array entry is set and the appropriate mark is drawn on the screen. From there, the checkgamewon method is called to see if this last move led to a win. If not, we call the checkallfilled method to see if all squares are now occupied, meaning the game ends in a tie. A hopefully more interesting use of two-dimensional arrays is to manage the bricks in a breakout game. See Example: Breakout This is a greatly-enhanced version of our last pong game, where there are now a series of bricks that get removed as the ball bounces off one of them. The most relevant to the current discussion is the two-dimensional array in the BrickCollection class. For the game as written, the bricks could be stored in a one-dimensional array or even an ArrayList. However, by storing them in a two-dimensional array, we can have the additional information about the positions if we wanted to enhance the game to have different point values based on position, for example. Two-Dimensional Matrices A very common use of two-dimensional arrays is the representation of matrices. We will look at an example of a class that represents two-dimensional square matrices and provides some basic operations on them. See Example: Matrix2D The class is capable of holding a square matrix of double values of any positive dimension. Comments within the example explain much of what is happening. Note in particular the use of the two-dimensional array as an instance variable which stores the matrix entries, the use of exceptions to handle error conditions, and the main method that tests out the methods of the class. 16